Revisiting AI and testing methods to infer FSM models of black-box systems

Groz, Roland; Simão, Adenilso da Silva; Brémond, Nicolas; Oriat, Catherine

doi:10.1145/3194733.3194736

Cited by 11 publications

(4 citation statements)

References 11 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…It predicts the behaviour of the virtual service. Groz et al [46] propose a method called hW‐inference, which can infer finite state machine (FSM) models from non‐resetting systems. To do so, they combined a learning‐based approach with conformance testing.…”

Section: Related Workmentioning

confidence: 99%

Towards automatically identifying the co‐change of production and test code

Huang,

Tang,

Chen

et al. 2024

Software Testing Verif & Rel

View full text Add to dashboard Cite

In software evolution, keeping the test code co‐change with the production code is important, because the outdated test code may not work and is ineffective in revealing faults in the production code. However, due to the tight development time, the production and test code may not be co‐changed immediately by developers. For example, we analysed the top 1003 popular Java projects on GitHub and found that nearly 9.3% of cases (i.e., 464,417) did not update their production and test code at the same time, that is, the production code is updated first, and then the test code is updated at intervals. The result indicates that much test code will not be updated in time. In this paper, we propose a novel approach, Jtup, to remind developers to co‐change the production code and test code in time. Specifically, we first define the co‐changed production and test code as a positive instance, while unchanged test code (i.e., production code changed and test code unchanged) as a negative instance. Then, we extract multidimensional features from the production code to characterize the possibility of their co‐change, including code change features, code complexity features, and code semantic features. Finally, several machine learning‐based methods are employed to identify the co‐changed production and test code. We conduct comprehensive experiments on 20 datasets, and the results show that the Accuracy, Precision, and Recall achieved by Jtup are 76.7%, 78.1%, and 77.4%, which outperforms the state‐of‐the‐art method.

show abstract

Section: Related Workmentioning

confidence: 99%

Towards automatically identifying the co‐change of production and test code

Huang,

Tang,

Chen

et al. 2024

Software Testing Verif & Rel

View full text Add to dashboard Cite

show abstract

“…Invariants can be used to augment state models [9,11]. Groz et al, use machine learning to heuristically infer state machine models of a un-resettable black-box system [27], however a significant difference between our method and theirs is that their method still relies on discrete events (such as HTTP request and responses) while our method does not assume that the input and outputs contain any kind of "events" happening at certain times. Our method aims to search for such events as change points in a continuous stream of data as time series.…”

Section: State Model Inferencementioning

confidence: 99%

Hybrid deep neural networks to infer state models of black-box systems

Mashhadi

Hemmati

2020

Proceedings of the 35th IEEE/ACM International Conference on Automated Software Engineering

View full text Add to dashboard Cite

Inferring behavior model of a running software system is quite useful for several automated software engineering tasks, such as program comprehension, anomaly detection, and testing. Most existing dynamic model inference techniques are white-box, i.e., they require source code to be instrumented to get run-time traces. However, in many systems, instrumenting the entire source code is not possible (e.g., when using black-box third-party libraries) or might be very costly. Unfortunately, most black-box techniques that detect states over time are either univariate, or make assumptions on the data distribution, or have limited power for learning over a long period of past behavior. To overcome the above issues, in this paper, we propose a hybrid deep neural network that accepts as input a set of time series, one per input/output signal of the system, and applies a set of convolutional and recurrent layers to learn the non-linear correlations between signals and the patterns, over time. We have applied our approach on a real UAV auto-pilot solution from our industry partner with half a million lines of C code. We ran 888 random recent system-level test cases and inferred states, over time. Our comparison with several traditional time series change point detection techniques showed that our approach improves their performance by up to 102%, in terms of finding state change points, measured by F1 score. We also showed that our state classification algorithm provides on average 90.45% F1 score, which improves traditional classification algorithms by up to 17%. CCS CONCEPTS• Computing methodologies → Neural networks; • Software and its engineering → Software reverse engineering; Requirements analysis.

show abstract

“…FSM is an artificial intelligence technique that allows predictability with a given set of inputs and a known current state; therefore, state transitions can be easily predicted and thus enable for easy testing. FSM also enable determination of reachability of a state thus when these states are represented in an abstract form; it is insentient clear whether one state can be arrived at from another state, and what is required to arrive the state [10].…”

Section: Design Of a Finite State Machine To Describe Normal And Anommentioning

confidence: 99%

Designing a Fused Machine Learning Model for the Provision of Smart Health Care in MANETS

Mindo¹,

Karume²,

Thiga³

2019

SJET

View full text Add to dashboard Cite

There is need to provide resilient security methodologies that do not require enormous computing resources. While entry prevention is the most viable disposition, it is not always possible to stop unauthorised access. Thus, it is critical to investigate the use of machine learning-based intrusion detection to buttress and provide sufficient security against DOS and other attacks in MANETs. Various anomaly-based intrusion detection systems employ varying techniques to identify anomalies in the context of diverse and valid variables. Most of these techniques, however, fail to capture and take account the physiognomies of MANETs. In the intervening time, usage of the internet of things in the provision of smart healthcare is expanding and the inherent risks snowballing. This study designed a model, which used a fusion of machine learning techniques through both simulation and a running prototype to achieve a more resilient intrusion detection system. The study was designed using functional decomposition methodology and implemented using PPDIO and evaluated on a MANET environment on both Linux NS 2 and further implemented on a network of Smart wearable devices and Raspberry Pi.

show abstract

Revisiting AI and testing methods to infer FSM models of black-box systems

Cited by 11 publications

References 11 publications

Towards automatically identifying the co‐change of production and test code

Towards automatically identifying the co‐change of production and test code

Hybrid deep neural networks to infer state models of black-box systems

Designing a Fused Machine Learning Model for the Provision of Smart Health Care in MANETS

Contact Info

Product

Resources

About